TransMAP Hub

Vessel Traffic Example¶

Using moving pandas with the TransMAP REST API

In [2]:

# Source ref: https://github.com/anitagraser/movingpandas-examples/blob/main/analysis-examples/2-ship-data.ipynb

make_url = lambda uri: "https://oak.cast.uark.edu/{0}".format(uri)

In [1]:

%matplotlib inline

In [3]:

from IPython.core.display import display, HTML
display(HTML("<style>.container { width:100% !important; }</style>"))

In [4]:

import requests
import numpy as np
import pandas as pd
import geopandas
from shapely.geometry import Point, LineString, Polygon
from datetime import datetime, timedelta
import matplotlib.pyplot as plt
import movingpandas as mpd

import warnings
warnings.simplefilter("ignore")

In [5]:

mpd.__version__
def make_date(year: int, month: int, day: int, hour: int = 0, minute: int = 0, second: int = 0):
    return datetime(year, month, day, hour, minute, second).strftime("%Y-%m-%dT%H:%M:%S")

In [12]:

ais_data = requests.get(make_url('ais/vessels'), params={
    "limit": 50000,
    "start_date": make_date(2020, 5, 29),
    "end_date": make_date(2020, 5, 30)
})
ais_data.raise_for_status()

In [13]:

df = pd.DataFrame(ais_data.json())
df

Out[13]:

	MMSI	BaseDateTime	LAT	LON	SOG	COG	Heading	VesselName	IMO	CallSign	VesselType	Status	Length	Width	Draft	Cargo	TranscieverClass	location
0	369330000	2020-05-29T00:00:00	29.13493	-90.20348	0.0	180.4	339.0	None	None	None	None	5.0	None	None	None	None	A	[-90.20348, 29.13493]
1	368076640	2020-05-29T00:00:00	29.60997	-84.58777	0.0	30.4	511.0	None	None	None	None	0.0	None	None	None	None	B	[-84.58777, 29.60997]
2	368043520	2020-05-29T00:00:00	27.88024	-97.27268	0.0	-80.6	244.0	None	None	None	None	0.0	None	None	None	None	B	[-97.27268, 27.88024]
3	367314740	2020-05-29T00:00:00	31.14192	-92.26107	1.2	-87.0	511.0	None	None	None	None	15.0	None	None	None	None	B	[-92.26107, 31.14192]
4	368926530	2020-05-29T00:00:00	29.82006	-91.54392	0.0	86.9	317.0	None	None	None	None	5.0	None	None	None	None	A	[-91.54392, 29.82006]
...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...
49995	367653590	2020-05-29T00:07:49	38.28540	-122.28518	1.1	44.6	511.0	NOMATA	None	WDH7990	37.0	None	18.0	5.0	None	None	B	[-122.28518, 38.2854]
49996	366926740	2020-05-29T00:07:49	33.75942	-118.27430	0.4	21.7	102.0	TIM QUIGG	IMO8987917	WDB6570	31.0	0.0	24.0	9.0	4.0	52.0	B	[-118.2743, 33.75942]
49997	366978690	2020-05-29T00:07:49	37.71104	-122.80402	6.7	-172.6	240.0	EAGLE	IMO8661654	WBP2187	31.0	0.0	29.0	9.0	None	None	A	[-122.80402, 37.71104]
49998	368011470	2020-05-29T00:07:49	29.62379	-89.90540	1.9	144.2	511.0	LOUISE	None	WDJ7929	31.0	0.0	17.0	7.0	None	31.0	B	[-89.9054, 29.62379]
49999	366878290	2020-05-29T00:07:49	39.27215	-74.98030	0.0	-153.3	511.0	RAM XV	IMO8767305	WDB2653	90.0	15.0	33.0	19.0	2.1	None	B	[-74.9803, 39.27215]

50000 rows × 18 columns

In [14]:

df2 = pd.DataFrame(df.location.tolist(), index= df.index)
gpdf = geopandas.GeoDataFrame(df, geometry=geopandas.points_from_xy(df2[0], df2[1]))

In [15]:

gpdf.set_crs('epsg:4326')

Out[15]:

	MMSI	BaseDateTime	LAT	LON	SOG	COG	Heading	VesselName	IMO	CallSign	VesselType	Status	Length	Width	Draft	Cargo	TranscieverClass	location	geometry
0	369330000	2020-05-29T00:00:00	29.13493	-90.20348	0.0	180.4	339.0	None	None	None	None	5.0	None	None	None	None	A	[-90.20348, 29.13493]	POINT (-90.20348 29.13493)
1	368076640	2020-05-29T00:00:00	29.60997	-84.58777	0.0	30.4	511.0	None	None	None	None	0.0	None	None	None	None	B	[-84.58777, 29.60997]	POINT (-84.58777 29.60997)
2	368043520	2020-05-29T00:00:00	27.88024	-97.27268	0.0	-80.6	244.0	None	None	None	None	0.0	None	None	None	None	B	[-97.27268, 27.88024]	POINT (-97.27268 27.88024)
3	367314740	2020-05-29T00:00:00	31.14192	-92.26107	1.2	-87.0	511.0	None	None	None	None	15.0	None	None	None	None	B	[-92.26107, 31.14192]	POINT (-92.26107 31.14192)
4	368926530	2020-05-29T00:00:00	29.82006	-91.54392	0.0	86.9	317.0	None	None	None	None	5.0	None	None	None	None	A	[-91.54392, 29.82006]	POINT (-91.54392 29.82006)
...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...
49995	367653590	2020-05-29T00:07:49	38.28540	-122.28518	1.1	44.6	511.0	NOMATA	None	WDH7990	37.0	None	18.0	5.0	None	None	B	[-122.28518, 38.2854]	POINT (-122.28518 38.28540)
49996	366926740	2020-05-29T00:07:49	33.75942	-118.27430	0.4	21.7	102.0	TIM QUIGG	IMO8987917	WDB6570	31.0	0.0	24.0	9.0	4.0	52.0	B	[-118.2743, 33.75942]	POINT (-118.27430 33.75942)
49997	366978690	2020-05-29T00:07:49	37.71104	-122.80402	6.7	-172.6	240.0	EAGLE	IMO8661654	WBP2187	31.0	0.0	29.0	9.0	None	None	A	[-122.80402, 37.71104]	POINT (-122.80402 37.71104)
49998	368011470	2020-05-29T00:07:49	29.62379	-89.90540	1.9	144.2	511.0	LOUISE	None	WDJ7929	31.0	0.0	17.0	7.0	None	31.0	B	[-89.9054, 29.62379]	POINT (-89.90540 29.62379)
49999	366878290	2020-05-29T00:07:49	39.27215	-74.98030	0.0	-153.3	511.0	RAM XV	IMO8767305	WDB2653	90.0	15.0	33.0	19.0	2.1	None	B	[-74.9803, 39.27215]	POINT (-74.98030 39.27215)

50000 rows × 19 columns

In [16]:

gpdf.plot()

Out[16]:

<AxesSubplot:>

In [17]:

gpdf['t'] = pd.to_datetime(gpdf['BaseDateTime'], format='%Y-%m-%dT%H:%M:%S')

In [18]:

gpdf = gpdf.set_index('t')

In [19]:

gpdf['SOG'].hist(bins=10, figsize=(15, 3))

Out[19]:

<AxesSubplot:>

In [20]:

print("Original size: {} rows".format(len(gpdf)))
gpdf = gpdf[gpdf.SOG>0]
print("Reduced to {} rows after removing 0 speed records".format(len(gpdf)))
gpdf['SOG'].hist(bins=10, figsize=(15,3))

Original size: 50000 rows
Reduced to 20940 rows after removing 0 speed records

Out[20]:

<AxesSubplot:>

In [21]:

gpdf['VesselType'] = gpdf['VesselType'].astype(float)
gpdf['VesselType'].value_counts().plot(kind='bar', figsize=(15,3))

Out[21]:

<AxesSubplot:>

In [22]:

%%time
MIN_LENGTH = 0.02 # degrees - May need to reproject to use meters
traj_collection = mpd.TrajectoryCollection(gpdf, 'MMSI', min_length=MIN_LENGTH)
print("Finished creating {} trajectories".format(len(traj_collection)))

Finished creating 550 trajectories
Wall time: 15.5 s

In [23]:

traj_collection = mpd.MinTimeDeltaGeneralizer(traj_collection).generalize(tolerance=timedelta(minutes=1))
# this is looking for a specific 't' index

In [24]:

shiptype_to_color = {'Passenger': 'blue', 'HSC': 'green', 'Tanker': 'red', 'Cargo': 'orange'}
# Should be changed to map to these codes: https://coast.noaa.gov/data/marinecadastre/ais/VesselTypeCodes2018.pdf
traj_collection.plot(column='VesselType', column_to_color=shiptype_to_color, linewidth=1, capstyle='round')

Out[24]:

<AxesSubplot:>

In [25]:

passenger = traj_collection
passenger.hvplot(title='Passenger ferries', line_width=2)

Out[25]:

In [ ]:

In [ ]:

---

Last update: 2021-05-04